UNIVERSITA’ DEGLI STUDI DI MESSINA
DIPARTIMENTO DI
XXIX CICLO DI DOTTORATO
SCIENZE BIOMEDICHE CLINICHE E SPERIMENTALI
Antipsychotic drug utilisation and safety in older persons with
dementia: an international pharmacoepidemiologic inquiry
TESI DI DOTTORATO:
Dott.sa Janet Sultana
UNIVERSITA’ DEGLI STUDI DI MESSINA
DIPARTIMENTO DI MEDICINA CLINICA E SPERIMENTALE
XXIX CICLO DI DOTTORATO IN
SCIENZE BIOMEDICHE CLINICHE E SPERIMENTALI
Antipsychotic drug utilisation and safety in older persons with
dementia: an international pharmacoepidemiologic inquiry
RELATORE:
Ch.mo Prof. Edoardo Spina
Febbraio 2016
UNIVERSITA’ DEGLI STUDI DI MESSINA
MEDICINA CLINICA E SPERIMENTALE
SCIENZE BIOMEDICHE CLINICHE E SPERIMENTALI
Antipsychotic drug utilisation and safety in older persons with
dementia: an international pharmacoepidemiologic inquiry
RELATORE:
Dedication
I would like to dedicate this dissertation to the few but important people in my life: to my father for his friendly and constant interest in my work; to my mother for her unquestioning belief in my ability to turn challenges into opportunities; to my sister for so rarely taking me seriously; to Shaun for providing an ironic but ultimately very effective stimulus to start writing; to Michael for providing many warm rays of sunshine.
Foreword
The present thesis is a collection of work carried out during the course of three years’ work at the Department of Clinical and Experimental Medicine, at the University of Messina. I have had the privilege to collaborate with several experts in the field of pharmacoepidemiology, pharmacology, geriatrics and statistics, and all of whom have enriched my understanding of the small part of “science” that I have worked on. As I dug deeper into my chosen field of research, I was increasingly humbled by how much there is to learn, including many lessons learned from my own errors. However, I find this somewhat encouraging in the light of the following words by Neils Bohr, the Danish 1922 recipient of the physics Nobel Prize: “An expert is a person who has made all the mistakes that can be made in a very narrow field.”
Acknowledgement
I am grateful to Prof. E. Spina for the opportunity to take part in the PhD programme in Biomedical, Clinical and Experimental Sciences and for the scientific freedom given to me. I am equally grateful to Dott. G. Trifiro’ for involving me in several pharmacoepidemiology studies, some of which form the basis of the present dissertation.
Table of Contents 1. Introduction
2. Reviews of safety concerns with antipsychotic use in dementia
2.1. Are the safety profiles of antipsychotic drugs used in dementia the same? An updated review of observational studies.
2.2. Antipsychotic use in elderly patients and the risk of pneumonia. 3. Drug utilisation studies
3.1. The Effect of Safety Warnings on Antipsychotic Drug Prescribing in Elderly Persons with Dementia in the United Kingdom and Italy: A Population-Based Study.
3.2. Antipsychotic use in dementia patients in a general practice setting: a Dutch population-based study.
4. Special topics in drug safety studies
4.1. Can information on functional and cognitive domains improve short-term mortality risk prediction among community-dwelling older persons? A population-based study using a UK primary care database
4.2. Drug safety warnings: a message in a bottle? 5. Conclusions
1. Introduction
The world is increasingly composed of an elderly population, bringing healthcare of older populations to the forefront. A number of older persons are privileged enough to enjoy healthy aging or have conditions that can be managed without significantly impacting their quality of life. However, many others are affected by dementia of various aetiologies, which leads to memory impairment and confusion. Often, this disorientation can give rise to psychological and behavioural symptoms of dementia (BPSD) including aggression, wandering and hallucinations. The only licensed pharmacological management of BPSD, specifically of aggression in dementia, is the short-term use of the antipsychotic risperidone. Nevertheless, several other antipsychotics are also used in dementia patients. The widespread use of antipsychotics in dementia, as well as concern about the serious adverse effects of these drugs and the lack of convincing data on their efficacy in BPSD has led to an awareness of the need to study antipsychotic drug utilisation and safety, in order to improve the appropriateness of drug prescribing.
The investigation of drug prescribing patterns as well as drug safety in the general population has been made possible on a large scale in past decade due the availability of electronic healthcare databases. There are several types of such databases, although these are primarily divided into electronic medical records (EMR) databases and electronic claims databases. Examples of EMR include The Health Improvement Database (THIN), a UK nationwide GP database, Health Search Database (HSD), an Italian nationwide GP database and the Integrated Primary Care Information database (IPCI), a Dutch nationwide database. These databases contain information that general practitioners register routinely during clinical practice. Unlike electronic medical records, claims data is registered when a person makes a claim for a medical service, drug prescription and so on. The claims are administrative data used for billing rather than for clinical purposes. Examples of claims databases include Arianna database (Caserta, in the Campania region of Italy), Medicare/Medicaid (samples of which are available for each state of the U.S.A., for groups of states or from all the U.S.A.) and the German Pharmacoepidemiological Research Database (Gepard), a German nationwide claims database.
The use of population-based data sources can provide a pharmacoepidemiologic perspective on drug utilisation and safety which has the following characteristics: it is a reflection of actual medical practice on a large scale which is often representative of a whole country; this permits researchers to capture information on potentially inappropriate prescribing practices and adverse outcomes without having to create ethical dilemmas. Further real-world aspects of drug utilisation such as low adherence, dose variations and medication switching can also often be captured with electronic healthcare databases, while they often cannot be captured using other study designs and data sources. There are other widely acknowledged advantages to using these data sources, such as the inclusion of data from persons who are not usually included in randomised clinical trials, such as frail elderly persons and persons with multi-morbidity and/or polypharmacy as well as the relatively long observation periods available for patients in population-based databases, compared to the much shorter observation periods available in prospective clinical studies. Other data sources and the associated study designs, which for the greater part consist of clinical trials, or other similar prospective studies, tend to reflect a somewhat artificial scenario as data is not collected during routine medical practice but on the basis of a protocol. As a result, these approaches are not suited for drug utilisation research. With regards to the evaluation of drug safety, clinical trials and other prospective study designs tend to have a short duration can therefore not be used to assess the long-term safety of drug use. This is a significant limitation in that several drugs are used chronically and associated adverse effects may take years to develop.
They may be able to capture short-term safety events, but clinical trials tend to be aimed at measuring effectiveness rather than safety as an outcome. As a result, such trials are not usually rich in short-term safety data. On the other hand, safety studies carried out using electronic healthcare databases are able to capture outcomes related to short-term and long-term drug exposure. In summary, observational studies carried out using electronic healthcare databases are currently the best tool in the public health inventory to investigate drug utilisation and safety. The aim of this thesis is to present a population-based pharmacoepidemiologic perspective on antipsychotic drug use and safety in dementia, through three approaches: a literature review on antipsychotic-associated safety outcomes in dementia, the implementation of drug utilisation studies using large general practice databases (THIN, HSD and IPCI) and a discussion of some specific topics of interest in the area, namely drug interactions in dementia patients, the role of frailty as a predictor of mortality and the influence of public health interventions on changing prescribing patterns and underlying risk of adverse events.
All the material presented in this thesis has been published in peer-reviewed
2. Reviews of safety concerns with antipsychotic use in dementia
The two papers presented in this chapter address the need for an updated review of observational studies investigating antipsychotic safety. The first paper in section 2.1. provides a comprehensive overview of several safety outcomes while the second paper in section 2.2. focuses on pneumonia as an outcome of antipsychotic use in the elderly.
2.1. Are the safety profiles of antipsychotic drugs used in dementia the same? An
updated review of observational studies. Trifiró G1, Sultana J1, Spina E1. Drug Saf. 2014
Jul;37(7):501-20.
1
Department of Clinical and Experimental Medicine, Section of Pharmacology, University of Messina, Messina, Italy
Abstract
With an increase in the global prevalence of dementia there is also an increase in behavioural and psychological symptoms of dementia (BPSD) for which antipsychotic drugs are often used. Despite several safety warnings on antipsychotic use in dementia as a class, there is little evidence to support the efficacy of antipsychotics in individual BPSD symptoms or to evaluate the drug safety profile by individual antipsychotic drug. There is emerging but scarce evidence that suggests an inter-drug variability between antipsychotic safety outcomes in BPSD. The objective of this review was to examine the existing literature on antipsychotic drug use in dementia patients, in particular to see whether inter-drug differences regarding antipsychotic safety were reported. A literature search was conducted for observational studies published in English from 2004-2014 that reported the risk of all-cause mortality, cerebrovascular events, pneumonia and other outcomes such as hip/femur fracture, deep vein thrombosis and hyperglycaemia. Six out of 16 mortality studies (38%), 7 out of 28 stroke studies (25%), 1 out of 6 pneumonia (17%) studies and 2 out of 6 fracture studies (33%) investigated inter-drug safety outcomes in elderly patients/dementia patients, while to our knowledge there are no studies investigating the inter-drug variation of deep-vein thrombosis and hyperglycaemia risk. The results of the observational studies provide mixed results on the safety of antipsychotics in BPSD but it is clear that there are differences between the safety profiles of antipsychotic drugs. Robust evidence of such inter-drug variability could significantly improve patient safety as antipsychotics become more targeted to the clinical risk factors.
Key points
• Despite increasing awareness of the safety issues surrounding antipsychotic drug use in BPSD, there is currently very limited in information on the inter-drug variation in risk as the vast majority of studies focus on all antipsychotics as a group or on atypical/conventional antipsychotics as a class.
• It is becoming apparent that there is indeed a difference between the risks associated with individual antipsychotic drugs in BPSD. Robust evidence of the risks associated with individual antipsychotic drugs could significantly improve the standards of clinical care by tailoring the specific therapeutic/safety properties to the clinical needs of individual patients.
1. Introduction
Globally, the estimated number of patients with dementia was 25 million in 2000 and is projected to rise to 63 million by 2030 [1].The clinical manifestations of dementia consist of cognitive and/or memory deterioration with progressive impairment of activities of daily living, as well as a variety of behavioural and psychological symptoms (BPSD)[2],[3]. These neuropsychiatric symptoms occur in more than 90% of patients with dementia and present a significant challenge for clinicians as well as care-givers[3]. BPSD is not a single behavior but comprises several symptoms such as agitation, psychosis and mood disorders, which usually co-occur and often recur. Patients with BPSD are more likely to need physical restraint, have a higher risk of early institutionalization and a higher risk of mortality [4],[5],[6]. In addition, BPSD negatively affects the quality of life of caregivers and other residents, if in a nursing home [7],[8]. The etiology of these symptoms still is not fully known.
Antipsychotics are often the first-line treatment for BPSD. They are generally distinguished as conventional (first-generation) or atypical (second-generation) antipsychotics. Conventional agents include butyrophenones (e.g. haloperidol), phenothiazines (e.g. chlorpromazine and thioridazine) and several others (e.g. indoles, thioxanthenes). Conventional antipsychotics were approved in the 1950’s mainly for the treatment of schizophrenia. Since then, these agents have also been used for the treatment of a broad spectrum of psychiatric disorders including BPSD despite a lack of scientific evidence supporting their use in dementia [9]. Currently, atypical antipsychotics include clozapine, risperidone, olanzapine, quetiapine, ziprasidone, aripiprazole, paliperidone, asenapine and amisulpride. The receptor binding profile among atypical antipsychotics differs substantially across different compounds such that these drugs cannot be truly considered a unique homogeneous therapeutic class [9-13]. Atypical antipsychotics were initially licensed in the 1990s and approved by the US Food and Drug Administration (FDA) exclusively for the treatment of schizophrenia. Nowadays, they are also approved for the treatment of bipolar mania, while their use in dementia has remained off-label. Only risperidone has been approved for the treatment of aggression in patients with Alzheimer’s disease in most European countries. Despite their off-label status in dementia, atypical antipsychotics have become the new standard of care for BPSD owing to their reported advantages over conventional agents, particularly a lower incidence of extrapyramidal symptoms (EPS) and tardive dyskinesia [9].In the late 1990s, atypical agents accounted for more than 80% of antipsychotic prescriptions in dementia in US nursing homes as well as in Canada[14]. In Europe, the use of atypical antipsychotics was lower even though it increased dramatically early after their introduction on the market[15].
1.1. Efficacy of antipsychotics in dementia
To date, more than 20 placebo controlled randomized clinical trials (RCTs) have investigated the efficacy of atypical antipsychotics for the treatment of BPSD, of which some were not published in full [21]. In their systematic review of 16 RCTs on atypical antipsychotics (olanzapine, quetiapine, risperidone, aripiprazole) for treatment of aggression, agitation and psychosis in dementia, Ballard and Waite concluded that risperidone and olanzapine have a modest efficacy in reducing aggression and psychosis, but both drugs were associated with serious adverse cerebrovascular events and extrapyramidal symptoms[21].Another meta-analysis of 7 RCTs of atypical antipsychotics (risperidone, olanzapine and quetiapine) reported neither a statistically nor a clinically significant difference in effectiveness as compared to placebo[22]. The findings from the meta-analyses were confirmed by a recent report of the CATIE-AD (Clinical Antipsychotic Trials of Intervention Effectiveness in Alzheimer's Disease) study which concluded that adverse effects
(olanzapine, risperidone, quetiapine versus placebo) offset the efficacy of atypical antipsychotic drugs for the treatment of psychosis, aggression, or agitation in patients with Alzheimer’s disease[23].More recently, a systematic review on the efficacy of atypical antipsychotics for off-label use in the treatment of elderly dementia patients with BPSD identified 14 placebo-controlled trials and assessed the efficacy of atypical antipsychotics using a total global outcome score including symptoms such as psychosis, mood alterations, and aggression. This systematic review reported small but statistically significant effect sizes ranging from 0.12 and 0.20 for aripiprazole, olanzapine, and risperidone, but indicated an absence of benefit with quetiapine[24].
In general, these randomized clinical trials (RCTs) had a short duration which may not reflect the use of antipsychotic drugs in clinical scenarios. In addition, placebo-controlled trial of 4 or more week’s duration may present significant potential for patient selection bias. This is because subjects with more severe psychosis or agitation may not choose to be enrolled into a placebo controlled trial as their symptoms are too severe to take the chance of getting a placebo treatment. This leaves more mildly psychotic or agitated subjected being enrolled into placebo-controlled trials resulting in an underestimation of the positive efficacy of these agents for psychosis and agitation. BPSD RCTs also present limitations when they include patients living at home, in senior independent or non-locked assisted living environments rather than patients living in locked assisted living or nursing home facilities. The latter are where the majority of antipsychotics are most commonly used and are where the most severe behaviors are encountered. A further limitation encountered in RCTs may be a lack of equivalence between the antipsychotic doses used, potentially favoring some drugs over others.
To our knowledge, at present no published data from double-blind RCTs on patients with dementia are available for amisulpride, clozapine, paliperidone, asenapine, andziprasidone, which are seldom or never used in BPSD[9].Overall, there is also very limited evidence of any benefit of antipsychotics in the treatment of BPSD over periods longer than 12 weeks, despite the fact that up to 60% of older people with dementia receive treatment with antipsychotics for more than 6 months[25].
1.2. Safety of antipsychotics in dementia
The safety profile of atypical and conventional antipsychotics has been questioned in recent years, as demonstrated by a number of warnings which have been issued by regulatory agencies [16]. Despite all safety warnings, recent studies document a persistent wide use of antipsychotics in dementia due to the lack of alternative pharmacological options. Valiyeva et al. demonstrated that the warnings slowed the growth in the use of atypical antipsychotics among patients with dementia, but they did not reduce the overall prescription rate of these drugs in Canada[17]. Similarly, other studies in USA and Europe observed a reduction in the use of atypical antipsychotics in dementia as a result of the initial safety alerts[18],[19],[20]. This decreasing trend was however counterbalanced by a switch towards conventional antipsychotics, even though these are reported to have a similar increase in the mortality risk[18],[19].For all these reasons, a re-evaluation of the possible risk minimization effects of the safety warnings as well as a thorough assessment of the long-term mortality of each single antipsychotic in dementia is much needed.
Various safety concerns have been encountered with antipsychotic use, including all-cause mortality, cardiac arrhythmias, peripheral vascular effects, metabolic effects, pneumonia, cerebrovascular accidents. Very little attention has however been given to the safety issues related to antipsychotic withdrawal in BPSD, an area that warrants further investigation particularly because the use of antipsychotics in BPSD is generally recommended in the short-term[26, 27].
The importance of studies targeting antipsychotic use in dementia patients is highlighted by age-related pharmacokinetic changes as well as potential drug-drug interactions that can result in higher and more variable drug concentrations in this population, thus further increasing the risk of toxicity [8],[28]. In addition, age-related pharmacodynamic changes generally require antipsychotic dose adjustment in elderly persons [29]. This is because the clinical effect of a drug is a function of the affinity with the target, the drug concentration at the site of action (depending on the ADME: absorption, distribution, metabolism, excretion) and patient characteristics such as age and sex[30]. Nevertheless, there are few such pharmacokinetic studies which assess the ADME parameters in dementia patients [28]. Drug metabolism and excretion may vary substantially in older persons and current clinical recommendations suggest prescribing only a quarter to half of the defined daily dose of antipsychotics in geriatric patients [30] in the absence of more detailed pharmacokinetic evidence.
In light of the wide use of antipsychotics in dementia patients as well as the uncertainty about their actual safety profile in clinical practice, we conducted an updated review of currently known safety issues of individual antipsychotics.
2. Methods
PubMed was searched for the following terms: ‘antipsychotics’, ‘antipsychotic drugs’, ‘antipsychotic agents’ and ‘mortality’, ‘all-cause mortality’, ‘death’ or ‘cerebrovascular events’, ‘cerebrovascular event’, ‘CVE’, ‘stroke’, ‘ischaemic stroke’, ‘ischemic stroke’, ‘haemorrhagic stroke’, ‘hemorrhagic stroke’, ‘transient ischaemic attack’, ‘transient ischemic attack’, ‘TIA’ or ‘pneumonia’, ‘community-acquired pneumonia’, ‘acute chest infections’, ‘bronchopneumonia’ or ‘hip fracture’, ‘femur fracture’, or ‘deep vein thrombosis’, ‘DVT’, or ‘hyperglycaemia’, ‘hyperglycemias’. Studies were included if they were observational, cohort, case–control or self-controlled studies published in English from 2004 to 2014. Studies were included irrespectively of whether the reference group was unexposed patients or not and with no restrictions related to diagnostic categories.
References of relevant original research as well as review articles were hand-searched to identify further studies. Two investigators (GT and JS) independently examined the titles and abstracts and obtained full texts of potentially relevant papers. Any disagreement was resolved through consensus. Information on study design, setting/data source, study population, outcomes measured, exposure and main findings (risk estimates where possible) were extracted for each study and tabulated. All confidence intervals reported were at the 95% level.
3. Results
3.1. All- cause mortality
In April 2005, the Food and Drug Administration (FDA) issued a warning to inform health professionals that the mortality rate among elderly patients with dementia-related behavioural disorders receiving an atypical antipsychotic was higher than that observed in placebo-treated patients [31]. One of the initial alarm triggers of antipsychotic safety was a pooled analysis of RCTs by the European Medicines Agency (EMA) in 2004, which reported a 2-fold increased risk of all cause-mortality with olanzapine compared to placebo[32]. A more extensive analysis was carried out by the Food and Drug Administration (FDA) shortly after, in 2005. The FDA meta-analysis included 17 RCTs which investigated all-cause mortality in olanzapine, risperidone, quetiapine and aripiprazole and reported a risk which was also approximately two-fold[31]. Further similar
analyses which arrived at similar conclusions were carried out in 2005. A meta-analysis of 5 olanzapine, 5 risperidone, 3 aripiprazole and 3 quetiapine trials which included only elderly dementia patients found that overall, all these drugs carried a risk of excess mortality [33]. In June 2008, the FDA stated that the conventional antipsychotics share a similar risk of increased mortality with the atypical antipsychotics [34].
Several subsequent observational studies investigated the risk of antipsychotic-related all-cause mortality in larger populations than RCTs and over longer exposure periods, and comparing the risk of atypical vs. conventional antipsychotics or non-use (Table 1). One such study was a cohort study published in the USA shortly after the FDA warning and which included 22,890 elderly atypical and conventional antipsychotic users (almost 50% with dementia) [35]. This study found a significant 30% increased risk of mortality with conventional antipsychotics compared to atypical antipsychotics. Similar results were found by a cohort study in British Colombia, namely a 26% increased risk of mortality within 180 days with conventional vs. atypical antipsychotics in elderly patients of whom, however, a lower proportion had dementia [36]. Another cohort study found that there was a 17% lower risk of mortality with atypical APs compared to conventional APs in dementia after 12 months[37]. None of these studies evaluated the risk of all-cause mortality associated with individual antipsychotics. The increasing evidence led the FDA to issue another warning in June 2008 on the high risk of mortality with conventional atypical antipsychotic use [38]. The risk of mortality was found to increase with increasing dose and was highest shortly after exposure[35],[36],[39],[40]. While the expanding research base has highlighted several relevant safety issues, several others such as the differential risk of mortality associated with individual antipsychotics remains unknown. A recent observational study by Huybrechts et al. suggested that the risk of mortality is differential, being highest for high dose haloperidol (high dose vs. low dose HR= 1.84 (1.38-2.43), with high and low dose defined using the median daily dose of chlorpromazine equivalent dose as a cut-off point) and lowest for low dose quetiapine (medium vs. low dose HR= 1.02 (0.89-1.18), with medium and low dose quetiapine defined as 575 and 0-50mg of chlorpromazine equivalent doses respectively)[41]. However, Huybrechts et al. investigated outcomes of only 5 antipsychotics (haloperidol, aripiprazole, olanzapine, quetiapine and ziprasidone, with risperidone as comparator), omitting several others. Two other recently published observational studies supports the finding that mortality risk varies by individual antipsychotic. One study reported that the highest mortality rates were for haloperidol (RR=1.54 (1.38–1.73)) while the lowest were for quetiapine (RR= 0.73 (0.67–0.80)).[42]The other study, which included only vascular dementia patients, reported higher but not statistically significant mortality rates for quetiapine (HR= 1.13 (0.92–1.37)) and lower mortality rates for risperidone (HR= 0.87 (0.60–1.27))[43].
The investigation of antipsychotic-related risk has also been carried out in more vulnerable sub-populations of dementia patients, such as those living in nursing homes[41],[44],[45],[46],[47],[48]. This is of particular importance as elderly persons living in nursing homes are likely to be more frail community-dwelling counterparts[49]. The comparative risk of specific cause mortality for individual antipsychotic agents is also poorly characterised, while more general comparisons between conventional and atypical antipsychotic classes have been investigated [50]. A recent study found a differential risk of specific causes of mortality in nursing homes, but included only one conventional antipsychotic compared to five atypical agent [41]. Huybrechts et al. found that haloperidol users in nursing homes had a higher risk of mortality compared with risperidone (HR=2.07 (1.89-2.26)) and quetiapine users had a lower mortality risk (RR=0.81 (0.75-0.88). It should be noted that some studies’ findings did not support this higher risk associated with conventional antipsychotics or even any antipsychotics in dementia, but such studies tended to
have very small populations and for this reason should be interpreted in the context of their limitations [51],[52].Several specific causes have been suggested to be at the root of antipsychotic mortality, including cerebrovascular events, pneumonia, peripheral vascular effects and metabolic effects, all of which are explored in further detail below.
A general consideration in these mortality studies and as well as other observational studies described below relates to the comparison group used. As can be seen from tables 1 to 4, the comparison group often consists of non-users of antipsychotics. This may result in the exposed and unexposed groups having important differences in dementia severity and overall frailty that increase the risk of death for exosposed groups independently of antipsychotic use alone.
3.2. Cerebrovascular effects
In October 2002, the marketing authorisation holder of risperidone notified all Canadian healthcare professionals that risperidone users had a higher rate of cerebrovascular events (CVEs) than placebo [53]. In March 2004, the UK Committee on Safety of Medicines recommended that health professionals avoid off-label use of atypical antipsychotics in elderly individuals with BPSD, particularly in those with a high baseline risk of stroke, [54] due to the observed CVE risk associated with these antipsychotics. At that time, information was reported only for olanzapine and risperidone although a similar alert was later released also by the manufacturer of aripiprazole[55].
Following the warning on antipsychotic-related stroke, several observational studies were conducted to compare the risk of stroke between atypical and conventional antipsychotics (Table 2). Most of these studies were conducted in an elderly population [50],[56],[57],[58],[59],[60], [61] but only two were restricted to older patients with dementia [45],[62]. Risperidone and olanzapine were associated with a nearly 3-fold increase in the risk of CVEs in dementia patients [47],[63],[64]. The risk of CVE was extrapolated through the whole class of drugs, despite concerns that this was unjustified [16]. Four observational studies suggest a higher risk of stroke with atypical than with conventional antipsychotics, even in dementia [58],[65],[66],[67]. However, at least four studies found that the risk of CVE was higher for conventional than atypical agents [16],[68],[69],[70]. Although there is very limited data comparing the risk of individual drugs within the class of atypical or conventional antipsychotics, a difference in CVE risk has been reported between phenothiazines (RR=5.79 (3.07–10.9)) and butyrophenones (RR=3.55 (1.56–8.07)) as compared to other atypicals (RR=2.46 (1.07–5.65)) compared to unexposed patients [60]. Another study reported that while conventional antipsychotics as a class were not associated with CVE, sulpiride was associated with CVE[71]. Similarly, as a class atypical antipsychotics were slightly associated with CVE while by individual agent, quetiapine and risperidone were not [71]. Although this study suggests that the difference in risk between class and individual drugs may be important, the results of this study must be interpreted with caution because risk estimates were not statistically significant.
Limited data is available on the dose-effect relationship between antipsychotic dose and stroke [16] although a recently published study reported that higher antipsychotic doses are associated with higher risks of CVE [66]. It was however reported that with regards to the temporal relationship between dose and effect, an elevated risk was found during the first weeks of treatment which decreases over time [72]. Another challenge of these studies was to identify unmeasured predictors of increased risk independently of the drug use. In one study, users of olanzapine and risperidone with several vascular risk factors (which were either not adequately treated or completely untreated)were more likely to develop CVE but it is unclear how much of this excess risk was due to the antipsychotics[73].There can be additional methodological concerns about the
diagnosis of CVE which may or may not be confirmed by radiological evidence. It is also not always clear how uniform or strict the definition of stroke or CVE employed across studies is, which may hinder direct comparison across studies.
The mechanism behind antipsychotic-related stroke in dementia is unknown but has been linked to orthostatic hypotension, hyperprolactinaemia resulting in atherosclerosis, thromboembolic events and excessive sedation [73],[74]. Some antipsychotic agents are known to antagonise alpha-adrenergic receptor which is a pathway for hypotension. Antipsychotic agents most likely to have a hypotensive effect are clozapine, quetiapine, risperidone and olanzapine in decreasing order, while haloperidol and ziprasidone were associated with the lowest risk of hypotension [16]. Atypical and conventional antipsychotics are both associated with venous thromboembolism (VTE) (see section on ‘Peripheral vascular effects’ below). On the other hand, it has also been hypothesised that the extrapyramidal effects of antipsychotics which lead to stiffness and sedation may later give rise to venous stasis and/or dehydration which could increase the risk of cerebrovascular events. Yet another putative mechanism is the thrombogenic effect due to hyperprolactinaemia which can result in enhance platelet reactivity [75].
Despite these suggested mechanisms, the association between stroke and antipsychotics was questioned because of the absence of a solid and proven biologically plausible explanation, uncertainty about the diagnostic accuracy of either TIA or stroke in the trials considered. The causal relationship between stroke and antipsychotics further questioned because patients were often affected by vascular dementia, which is itself associated with cerebrovascular risk. Cognitive impairment and stroke are very much related and older patients with Alzheimer’s disease are more likely to die from cerebrovascular disease than non-demented elderly subjects [16]. Following the warnings on antipsychotic-related stroke, several observational studies were conducted to compare the risk of stroke between atypical and conventional antipsychotics. Most of these studies were conducted in elderly populations[50],[56],[57],[58],[59],[60],[61], but only two were restricted to older patients with dementia [45],[62]. The study by Gill et al. reported that long-term care resident status was a risk factor for CVE (RR= 1.15 (0.82-1.6)) for atypical antipsychotics, with conventional antipsychotics being the referent), as was a history of atrial fibrillation factor (RR= 1.23 (0.70-2.02)) [62]. This study did not investigate the differential risk associated with antipsychotic use and provided no information on the duration or dose of antipsychotics used. Liperoti et al. reported the dose only descriptively without investigating the association between different doses and risk of CVE, although they provide risk estimates for the hospitalization of elderly nursing home residents with a diagnosis of stroke or transient ischemic attack for two specific atypical antipsychotics: risperidone versus no use (OR =0.87 (0.67-1.12)) and olanzapine versus no use (OR= 1.32 (0.83-2.11)) [45]. Liperoti et al. also found that a history of CVE was an effect modifier for atypical antipsychotic use (RR = 4.63 (1.35-32.63)), in particular for olanzapine use (RR= 3.71 (1.55-8.84)) and to a lesser degree for risperidone (RR=1.49 (0.93-2.38). However, no other antipsychotics were considered individually, whether atypical or conventional [45].
3.3. Pneumonia
Infections, primarily pneumonia, have been listed as one of the most prevalent causes of death among elderly demented patients using antipsychotics, both in clinical trials and observational studies [16](Table 3). It is difficult to explore the relationship between antipsychotics and pneumonia since patients with dementia already have a higher risk of aspiration pneumonia, which makes any observational study liable to confounding by indication. Moreover, frail older patients may initially manifest pneumonia with delirium requiring antipsychotic drug treatment, thus also raising the potential for protopathic bias in observational studies [76]. A Dutch study investigating
the association between the hospital-based diagnosis of pneumonia and antipsychotic use reported a 3-fold increased risk during use of atypical and 1.6-fold increase during use of conventional antipsychotics as compared to non-use in an elderly population [76]. Setoguchi et al. found a slightly higher rate of fatal pneumonia during conventional antipsychotic use relative to atypical antipsychotic use, but the overall risk of antipsychotic linked to the use was not increased compared to non-use in a cohort of elderly patients [50]. Trifirò et al. showed that the use of either atypical or conventional antipsychotics in elderly patients is associated with an increase in the risk of pneumonia in a dose-dependent manner [77]. Looking at individual agents, Trifirò et al. found the highest risk of pneumonia was associated with risperidone (OR= 3.51 (1.94–6.36)), followed by zuclopenthixol (OR= 2.25 (1.00–5.08)), haloperidol (OR= 1.95 (1.20–3.17)), olanzapine (OR =1.90 (0.61–5.90)) and paliperidone (OR =1.55 (1.00–2.43)). Given the frequency and poor prognosis of pneumonia in elderly dementia patients, it is important to explore the relationship between use of each single antipsychotic and pneumonia in dementia patients. This has only been explored by one study, to our knowledge, a recently published study which found that, using risperidone as comparator, olanzapine and ziprasidone had a stronger association with pneumonia than quetiapine and aripiprazole; this study was however limited by the small numbers of ziprasidone and aripiprazole users [67].
The possible mechanisms of antipsychotic-induced pneumonia remain speculative. It is likely that antipsychotics may induce aspiration pneumonia in dementia patients through many possible mechanisms involving extrapyramidal adverse events, dysphagia, or sedation, as a result of modulation of dopamine, cholinergic, and H1-histaminergic receptors, respectively [78]. Due to differences in the receptor binding profiles among various antipsychotics, the risk of pneumonia for any single antipsychotic and the underlying mechanism should be further investigated. It has already been shown in one study that the risk of pneumonia is differential between atypical (OR= 5.97 (1.49–23.98)) and conventional antipsychotics (OR= 1.71 (0.76–3.87)), between subclasses such as butyrophenones (OR= 1.42 (0.59–3.37)) and other antipsychotics such as thioxanthene, diphenylbutylpiperidine, and benzamide derivatives (OR= 2.84 (0.74–10.92)) as well as between some individual antipsychotics (see above) [78]. However, this study only considered five antipsychotics individually, leaving doubts about the risk associated with other antipsychotics. In addition, methodological issues such as confounding by indication and protopathic bias obscure the association between antipsychotic drugs and pneumonia and their effect on risk estimates must be considered thoroughly in order to avoid misleading results [78].
3.4. Cardiac arrhythmias
Since the early sixties, sudden cardiac death (SCD) has been reported with conventional antipsychotic use, mostly haloperidol and thioridazine [79]. Particular attention has been paid to the ability of antipsychotic drugs to prolong the QTc interval which may result in Torsade de Pointes and other potentially fatal ventricular arrhythmias[80]. QTc interval prolongation was reported to be highest for thioridazine, sertindole, pimozide, haloperidol, quetiapine and ziprasidone, in decreasing order [81]. Several observational studies have confirmed the signals from spontaneous reports suggesting that conventional antipsychotics are associated with an increased risk of SCD[79],[82],[83],[84]. In particular, thioridazine was withdrawn from the market in some countries due to concerns of cardiac arrhythmia [85]. Recently two large US studies found that the risk of SCD is increased also with the four most frequently prescribed atypical antipsychotics (clozapine, olanzapine, quetiapine and risperidone)[44],[82],[84]. On the basis of this evidence, electrocardiography monitoring would be prudent in routine clinical care if antipsychotics are prescribed to elderly patients[86, 87]. No studies investigated arrhythmogenic potential of antipsychotics in patients with dementia, specifically, and none of the currently available studies
had the statistical power to look at dose and duration effects of individual drugs on the SCD risk. 3.5. Peripheral vascular effects
Antipsychotic use has been associated with the occurrence of venous thromboembolism (VTE), an association that was recently reviewed by the UK Medicines and Healthcare products Regulatory Agency (MHRA)[88]. A relationship between antipsychotic medications and VTE was first suggested around five decades ago[88]. However, despite early descriptions and subsequent reports of VTE associated with antipsychotic use, evidence for a true link has not been clearly established. Reviews of the available data for aripiprazole, clozapine and olanzapine have led to warnings about VTE being added to their Summaries of Product Characteristics (SPCs)[88]. There are now several studies on VTE and antipsychotics[89],[90],[91],[92],[93],[84], [94],[95],[96],[97],[98],[99], [100]mostly on young patients with schizophrenia and with methodological limitations (small sample size, inadequate control of confounding). Findings about specific drugs were inconsistent, but all studies concluded that an increased risk of VTE with atypical and/or conventional antipsychotics was likely[88]. Very little data is available on the peripheral vascular effects of antipsychotics in dementia, which is highly relevant given the extensive use of other potentially interacting medications acting on serotonin receptors and platelet function in these patients. Only one study investigating the link between antipsychotic use and VTE has been published to our knowledge, a recent nested case-control using the a cohort of 72,591 dementia patients[101]. This study found that among users of antipsychotics from this population, current users had a statistically significant increased risk of VTE (OR= 1.23 (1.01-1.60)) compared to controls, defined as dementia patients at risk of VTE. Within the subgroup of current antipsychotic users, new users had a higher risk of VTE than controls, than prevalent users or than past users. The risk of VTE did not appear to vary between first and second generation antipsychotics when these were analysed as separate groups; the risk of VTE associated with individual antipsychotics was not investigated [101]. No other studies investigate the differential risk of VTE associated with individual antipsychotics.
3.6. Metabolic effects
While mortality, stroke and pneumonia were the main focus of research, several other adverse events related to antipsychotic use are also a source of concern (Table 4). Metabolic effects of antipsychotics are a long-term safety concern and may contribute to further increase the cardiovascular risk in older people with dementia[102]. In patients with either schizophrenia or bipolar disorder, the use of antipsychotics (i.e. olanzapine, clozapine) has been associated with metabolic abnormalities including weight gain, lipid disturbances and altered glucose homeostasis[103]. Whether elderly patients with BPSD receiving antipsychotics develop similar disturbances is still unclear. Metabolic effects of antipsychotics in elderly patients with dementia are difficult to assess in general, as food intake is reduced in these subjects. Only a few and relatively small studies have been published so far on this association. Rondanelli et al. concluded, on the basis of 36 nursing home residents with AD, that treatment with low-dose atypical antipsychotics does not lead to weight gain or increase in the risk of type II diabetes or lipid metabolism abnormalities[104]. In contrast, the CATIE-AD trial reported weight gain during use of olanzapine, quetiapine and risperidone in 421 AD patients and the risk increased over time[105]. Beside an increase in body weight, there was no apparent effect on glucose levels, total cholesterol and triglycerides levels, apart from an unfavourable change in HDL cholesterol and girth with olanzapine. Post-hoc analyses of other studies with olanzapine and risperidone were consistent with the CATIE-AD trial[16]. A recently published Canadian study by Lipscombe et al., carried out using four administrative databases in Ontario, found that among older patients with
diabetes, the initiation of treatment with antipsychotic drugs was associated with an increased risk hyperglycemias [106].
The risk of hyperglycaemia appeared to be much higher for incident (RR= 15.4(8.12-29.2)) use than prevalent (RR=1.36(1.03-1.79)) antipsychotic use for insulin-treated patients taking any antipsychotic, and slightly higher for atypical AP use (RR=1.4(1.06-1.85)) than for conventional AP use (RR=1.27(0.75-2.12)) among these patients. The overall risk of hyperglycaemia was slightly lower when patients were prescribed oral hypoglycaemic agents and were incident antipsychotic users (RR=14.4(8.71-23.8)) compared to when they were prescribed insulin. Lipscombe et
al.however, did not investigate the risk of hyperglycaemia associated with individual antipsychotics and this remains unknown at present[106]. At present, the association between antipsychotic use and either hyperglycaemia in elderly diabetic patients with dementia or new onset diabetes in elderly dementia patients requires better investigation[16]. In addition, it should be clarified if such possible metabolic effects (i.e. hyperglycaemia, hypercholesterolemia, hypertriglyceridemia and weight gain) of antipsychotics lead to a clinically relevant increased risk of all-cause mortality in these patients over the period of antipsychotic treatment in dementia patients
4. Directions for future research on antipsychotics in BPSD
The body of scientific evidence regarding the safety and efficacy of antipsychotics in BPSD is expanding; however there are several significant research gaps that still exist. There is very limited data on the safety of individual antipsychotics, as illustrated above. In addition, most antipsychotic safety studies tend to group all BPSD patients together rather than evaluating outcomes by individual BPSD symptoms. Only one study has investigated an outcome (mortality) considering symptoms such as delirium and hallucinations in dementia patients prescribed antipsychotics[107]. The type of dementia associated with BPSD is also likely to influence the safety of antipsychotics, but this has been a somewhat neglected area of clinical research.
So far, comprehensive safety data about long-term use of antipsychotics in dementia patients in various settings and different European countries is missing[108]. Although various clinical trials and observational studies have investigated the post-marketing risk of all-cause mortality [35],[36],[37, 40], [43],[44],[51],[109], cerebrovascular adverse events [45],[50],[56],[57],[58],[59],[60],[61], [62]sudden cardiac death[79],[80],[82],[83],[84], venous thromboembolism/pulmonary embolism[88],[89],[90],[91],[92],[93],[94],[95],[96],[97],[98],[99], diabetes mellitus and other metabolic effects [76],[102],[103],[104],[105] and community-acquired pneumonia[76], [77], [78]associated with atypical and conventional antipsychotics, few of these studies were able to properly assess the short- and long-term risk for each single antipsychotic separately in a well-powered study, despite emerging evidence that their clinical characteristics seem to be different. Of the 16 mortality studies we considered in this review, only 6 investigated the risk of some individual antipsychotics [36],[37],[41],[42], [43], ,[48],[100]; of the 18 stroke studies only 7 investigated individual antipsychotics [45],[56],[57],[61], [71],[67],[110]. Only 1 out of 6 studies investigated the risk of pneumonia with individual antipsychotics [111]; only 2 out of 6 studies investigated the risk of hip fracture with individual antipsychotics [112],[113] and neither of the studies investigating the risk of DVT and hyperglycaemia evaluated the individual risk of antipsychotics. This missing information could have important implications for choosing the drug of least risk in populations particularly prone to specific ADRs. The optimal dose associated with least risk of various ADRs is also not well-investigated with regards to antipsychotic use in BPSD, a potentially important aspect of antipsychotic safety given that the dose of APs in BPSD is lower than that in schizophrenia or bipolar disease. This is particularly relevant and factoring given the pharmacokinetic and pharmacodynamic characteristics of dementia patients. Most observational
studies were focused on elderly populations rather than elderly dementia patients specifically. Moreover, these studies were conducted in a specific region or country (mostly USA), which restricts heterogeneity in exposure, thus resulting in lack of statistical power to evaluate the entire range of individual antipsychotics and prevent generalizability of the findings to dementia patients from other Countries. For instance, results from US observational studies can be hardly generalized to the European setting due to the differences about the prescribing pattern of antipsychotics in dementia between USA and Europe.
Individual RCTs were powered on efficacy outcomes and could not provide useful insights on safety outcomes. In addition, systematic reviews and meta-analyses of randomized data were not able to disentangle the absolute and relative risk of each antipsychotic versus placebo and versus other antipsychotics. For some newer atypical antipsychotics, findings have not been systematically reviewed yet. Furthermore, the safety of antipsychotics in BPSD is rarely compared to other off-label medications, such that the risks cannot be compared to other therapeutic options. The long-term safety of antipsychotics in BPSD in particular presents a critical limitation in BPSD research so far, as there is very limited evidence of any benefit of these drugs for the treatment of BPSD over periods longer than 12 weeks. Most dementia patients discontinue antipsychotic treatment after a few weeks, yet a relevant proportion of them take these drugs for much longer periods. The AGIT and DART studies did not demonstrate any advantage for antipsychotics compared to placebo over six months[114],[115],[116],[117], and the CATIE study described no overall benefit[22]. However the CATIE trial did indicate that antipsychotics were less likely to be discontinued because of perceived ineffectiveness over nine months than placebo[22]. Furthermore, there is very limited data comparing antipsychotics to other off-label drugs in BPSD and similarly, limited data on the withdrawal of individual antipsychotic agents.
5. Conclusion
There are few observational studies that report the risk of adverse events with individual antipsychotics in elderly dementia patients. The highest risk of mortality was reported for haloperidol [36],[39] and chlorpromazine [48] while the lowest risk was reported for olanzapine [36], quetiapine [41], [42] and ziprasidone [41]. The evidence is much less clear-cut for stroke, with some studies reporting an increased [56],[61] or decreased [45],[71], [67],[110] risk with risperidone, increased [61],[110], or decreased [57],[71] risk with quetiapine and increased [57] or decreased [110] risk with haloperidol. Only one study investigated the risk pneumonia with individual antipsychotics but this did not provide a risk estimate nor was it sufficiently powered [111]. The risk of fracture was highest for zuclopenthixol [113] and haloperidol [112],[113] although too few studies investigated this outcome for these results to be conclusive. Only one study investigated DVT [101] and hyperglycaemia [106] respectively, neither of which considered the individual risk of antipsychotics.
While research on antipsychotic efficacy and safety in BPSD has expanded, research on the efficacy of individual antipsychotics in specific BPSD symptoms and the safety issues of individual antipsychotic use in BPSD has lagged behind. There are several studies suggesting a difference between the safety profile of atypical and conventional antipsychotics but there are only few studies on individual antipsychotic safety, suggesting that inter-drug difference in this respect are indeed being over-looked.
References
1. Ferri CP, Prince M, Brayne C, Brodaty H, Fratiglioni L, Ganguli M, et al. Global prevalence of dementia: a Delphi consensus study. Lancet. 2005 Dec 17;366(9503):2112-7.
2. Burns A, Iliffe S. Dementia. Bmj. 2009;338:b75.
3. Ballard C, Creese B, Corbett A, Aarsland D. Atypical antipsychotics for the treatment of behavioral and psychological symptoms in dementia, with a particular focus on longer term outcomes and mortality. Expert Opin Drug Saf. 2011 10(1):35-43.
4. Testad I, Aasland AM, Aarsland D. The effect of staff training on the use of restraint in dementia: a single-blind randomised controlled trial. International journal of geriatric psychiatry. 2005 Jun;20(6):587-90.
5. Selbaek G, Kirkevold O, Engedal K. The course of psychiatric and behavioral symptoms and the use of psychotropic medication in patients with dementia in Norwegian nursing homes--a 12-month follow-up study. Am J Geriatr Psychiatry. 2008 Jul;16(7):528-36.
6. Walsh JS, Welch HG, Larson EB. Survival of outpatients with Alzheimer-type dementia. Ann Intern Med. 1990 Sep 15;113(6):429-34.
7. Rosdinom R, Zarina MZ, Zanariah MS, Marhani M, Suzaily W. Behavioural and psychological symptoms of dementia, cognitive impairment and caregiver burden in patients with dementia. Prev Med. Jan 8.
8. Corbett A, Smith J, Creese B, Ballard C. Treatment of behavioral and psychological symptoms of Alzheimer's disease. Curr Treat Options Neurol. 2012 2012 14(2):113-25.
9. Liperoti R, Pedone C, Corsonello A. Antipsychotics for the treatment of behavioral and psychological symptoms of dementia (BPSD). Current neuropharmacology. 2008 Jun;6(2):117-24. 10. Factor SA. Pharmacology of atypical antipsychotics. Clin Neuropharmacol. 2002 May-Jun;25(3):153-7.
11. Shahid M, Walker GB, Zorn SH, Wong EH. Asenapine: a novel psychopharmacologic agent with a unique human receptor signature. J Psychopharmacol. 2009 Jan;23(1):65-73.
12. Burris KD, Molski TF, Xu C, Ryan E, Tottori K, Kikuchi T, et al. Aripiprazole, a novel antipsychotic, is a high-affinity partial agonist at human dopamine D2 receptors. J Pharmacol Exp Ther. 2002 Jul;302(1):381-9.
13. Stahl S. Antipsychotic agents. In: Stahl SM, editor. Stahl’s Essential Psychopharmacology. 4th Ed. ed. New York: Cambridge University Press; 2013.
14. Liperoti R, Mor V, Lapane KL, Pedone C, Gambassi G, Bernabei R. The use of atypical antipsychotics in nursing homes. J Clin Psychiatry. 2003 Sep;64(9):1106-12.
15. Trifiro G, Spina E, Brignoli O, Sessa E, Caputi AP, Mazzaglia G. Antipsychotic prescribing pattern among Italian general practitioners: a population-based study during the years 1999-2002. Eur J Clin Pharmacol. 2005 Mar;61(1):47-53.
16. Trifirò G, Spina E, Gambassi G. Use of antipsychotics in elderly patients with dementia: do atypical and conventional agents have a similar safety profile? Pharmacol Res. 2009;59(1-12). 17. Valiyeva E, Herrmann N, Rochon PA, Gill SS, Anderson GM. Effect of regulatory warnings on antipsychotic prescription rates among elderly patients with dementia: a population-based time-series analysis. Cmaj. 2008 Aug 26;179(5):438-46.
18. Trifiro G, Sini G, Sturkenboom MC, Vanacore N, Mazzaglia G, Caputi AP, et al. Prescribing pattern of antipsychotic drugs in the Italian general population 2000-2005: a focus on elderly with dementia. Int Clin Psychopharmacol. 2010 Jan;25(1):22-8.
19. Dorsey ER, Rabbani A, Gallagher SA, Conti RM, Alexander GC. Impact of FDA black box advisory on antipsychotic medication use. Arch Intern Med. 2010 Jan 11;170(1):96-103.
20. Kales HC, Zivin K, Kim HM, Valenstein M, Chiang C, Ignacio RV, et al. Trends in antipsychotic use in dementia 1999-2007. Arch Gen Psychiatry. 2011 Feb;68(2):190-7.
21. Ballard C, Waite J. The effectiveness of atypical antipsychotics for the treatment of aggression and psychosis in Alzheimer's disease. Cochrane Database Syst Rev. 2006(1):CD003476.
22. Yury CA, Fisher JE. Meta-analysis of the effectiveness of atypical antipsychotics for the treatment of behavioural problems in persons with dementia. Psychother Psychosom. 2007;76(4):213-8.
23. Schneider LS, Tariot PN, Dagerman KS, Davis SM, Hsiao JK, Ismail MS, et al. Effectiveness of atypical antipsychotic drugs in patients with Alzheimer's disease. N Engl J Med. 2006 Oct 12;355(15):1525-38.
24. Maher AR, Maglione M, Bagley S, Suttorp M, Hu JH, Ewing B, et al. Efficacy and comparative effectiveness of atypical antipsychotic medications for off-label uses in adults: a systematic review and meta-analysis. Jama. 2011 Sep 28;306(12):1359-69.
25. Barnes TR, Banerjee S, Collins N, Treloar A, McIntyre SM, Paton C. Antipsychotics in dementia: prevalence and quality of antipsychotic drug prescribing in UK mental health services. Br J Psychiatry. 2012 Sep;201(3):221-6.
26. Ballard C, Howard R. Neuroleptic drugs in dementia: benefits and harm. Nature reviews. 2006 Jun;7(6):492-500.
27. National Institute of Health and Care Excellence. Dementia: Supporting people with dementia and their carers in health and social care. 2006 [cited 16.04.14]; Available from: http://pathways.nice.org.uk/pathways/dementia#path=view%3A/pathways/dementia/dementia- interventions.xml&content=view-node%3Anodes-pharmacological-interventions-for-non-cognitive-symptoms
28. Pollock B, Forsyth C, Bies R. The critical role of clinical pharmacology in geriatric psychopharmacology. Clinical pharmacology and therapeutics. 2009 Jan;85(1):89-93.
29. Trifiro G, Spina E. Age-related changes in pharmacodynamics: focus on drugs acting on central nervous and cardiovascular systems. Curr Drug Metab. 2011 Sep;12(7):611-20.
30. [No authors listed]. A roadmap to key pharmacologic principles in using antipsychotics. Prim Care Companion J Clin Psychiatry. 2007;9(6):444-54.
31. Food and Drug Association. Public Health Advisory: Deaths with Antipsychotics in Elderly Patients with Behavioral Disturbances. 2005.
32. European Medicines Agency (EMEA). EMEA public statement on the safety of olanzapine(Zyprexa, Zyprexa Velotab). 2004 [cited 18.09.2013]; Available from:
http://www.bfarm.de/SharedDocs/1_Downloads/DE/Pharmakovigilanz/am-sicher-akt/emea-zyprexa-st.pdf?__blob=publicationFile
33. Schneider LS, Dagerman KS, Insel P. Risk of death with atypical antipsychotic drug treatment for dementia: meta-analysis of randomized placebo-controlled trials. Jama. 2005 Oct 19;294(15):1934-43.
34. Food and Drug Association. Information on Conventional Antipsychotics. 2008.
35. Wang PS, Schneeweiss S, Avorn J, Fischer MA, Mogun H, Solomon DH, et al. Risk of death in elderly users of conventional vs. atypical antipsychotic medications. N Engl J Med. 2005 Dec 1;353(22):2335-41.
36. Schneeweiss S, Setoguchi S, Brookhart A, Dormuth C, Wang PS. Risk of death associated with the use of conventional versus atypical antipsychotic drugs among elderly patients. Cmaj. 2007 Feb 27;176(5):627-32.
37. Kales HC, Valenstein M, Kim HM, McCarthy JF, Ganoczy D, Cunningham F, et al. Mortality risk in patients with dementia treated with antipsychotics versus other psychiatric medications. The American journal of psychiatry. 2007 Oct;164(10):1568-76; quiz 623.
38. Food and Drug Association (FDA). Information for Healthcare Professionals: Conventional
Antipsychotics. . 2008 [cited 19.09.2013]; Available from:
http://www.fda.gov/Drugs/DrugSafety/PostmarketDrugSafetyInformationforPatientsandProviders/uc m124830.htm
39. Hollis J, Grayson D, Forrester L, Brodaty H, Touyz S, Cumming R. Antipsychotic medication dispensing and risk of death in veterans and war widows 65 years and older. Am J Geriatr Psychiatry. 2007 Nov;15(11):932-41.
40. Gill SS, Bronskill SE, Normand SL, Anderson GM, Sykora K, Lam K, et al. Antipsychotic drug use and mortality in older adults with dementia. Ann Intern Med. 2007 Jun 5;146(11):775-86. 41. Huybrechts KF, Gerhard T, Crystal S, Olfson M, Avorn J, Levin R, et al. Differential risk of death in older residents in nursing homes prescribed specific antipsychotic drugs: population based cohort study. Bmj. 2012;344:e977.
42. Kales HC, Kim HM, Zivin K, Valenstein M, Seyfried LS, Chiang C, et al. Risk of mortality among individual antipsychotics in patients with dementia. The American journal of psychiatry. 2012 Jan;169(1):71-9.
43. Sultana J, Chang CK, Hayes RD, Broadbent M, Stewart R, Corbett A, et al. Associations between risk of mortality and atypical antipsychotic use in vascular dementia: a clinical cohort study. International journal of geriatric psychiatry. 2014 Mar 14.
44. Liperoti R, Onder G, Landi F, Lapane KL, Mor V, Bernabei R, et al. All-cause mortality associated with atypical and conventional antipsychotics among nursing home residents with dementia: a retrospective cohort study. J Clin Psychiatry. 2009 Oct;70(10):1340-7.
45. Liperoti R, Gambassi G, Lapane KL, Chiang C, Pedone C, Mor V, et al. Cerebrovascular events among elderly nursing home patients treated with conventional or atypical antipsychotics. J Clin Psychiatry. 2005 Sep;66(9):1090-6.
46. Huybrechts KF, Rothman KJ, Silliman RA, Brookhart MA, Schneeweiss S. Risk of death and hospital admission for major medical events after initiation of psychotropic medications in older adults admitted to nursing homes. Cmaj. 2011 Apr 19;183(7):E411-9.
47. De Deyn PP, Katz IR, Brodaty H, Lyons B, Greenspan A, Burns A. Management of agitation, aggression, and psychosis associated with dementia: a pooled analysis including three randomized, placebo-controlled double-blind trials in nursing home residents treated with risperidone. Clin Neurol Neurosurg. 2005 Oct;107(6):497-508.
48. Hollis J, Forrester L, Brodaty H, Touyz S, Cumming R, Grayson D. Risk of death associated with antipsychotic drug dispensing in residential aged care facilities. Aust N Z J Psychiatry. 2007 Sep;41(9):751-8.
49. Aparasu RR, Chatterjee S, Mehta S, Chen H. Risk of death in dual-eligible nursing home residents using typical or atypical antipsychotic agents. Med Care. 2012 Nov;50(11):961-9.
50. Setoguchi S, Wang PS, Alan Brookhart M, Canning CF, Kaci L, Schneeweiss S. Potential causes of higher mortality in elderly users of conventional and atypical antipsychotic medications. J Am Geriatr Soc. 2008 Sep;56(9):1644-50.
51. Raivio MM, Laurila JV, Strandberg TE, Tilvis RS, Pitkala KH. Neither atypical nor conventional antipsychotics increase mortality or hospital admissions among elderly patients with dementia: a two-year prospective study. Am J Geriatr Psychiatry. 2007 May;15(5):416-24.
52. Suh GH, Shah A. Effect of antipsychotics on mortality in elderly patients with dementia: a 1-year prospective study in a nursing home. Int Psychogeriatr. 2005 Sep;17(3):429-41.
53. Janssen-Ortho. Risperdal (risperidone) and cerebrovascular adverse events in placebo-controlled dementia trials [Dear Healthcare Professional Letter]. Toronto: Janssen–Ortho Inc.; 2002 Oct 11. Available: www .hc-sc.gc.ca/hpb-dgps/therapeut/zfiles/english /advisory /industry/risperdal1_e.pdf 2002 [cited; Available from:
54. Committee on Safety of Medicines, . Latest News: Atypical antipsychotic drugs and stroke. . 9 March 2004.
55. Italian Drug Agency. Nota informativa importante concordata con le autorità regolatorie europee e l’AIFA. .Abilify (aripiprazolo). . 2005.
56. Herrmann N, Mamdani M, Lanctot KL. Atypical antipsychotics and risk of cerebrovascular accidents. The American journal of psychiatry. 2004 Jun;161(6):1113-5.
57. Finkel S, Kozma C, Long S, Greenspan A, Mahmoud R, Baser O, et al. Risperidone treatment in elderly patients with dementia: relative risk of cerebrovascular events versus other antipsychotics. Int Psychogeriatr. 2005 Dec;17(4):617-29.
58. Percudani M, Barbui C, Fortino I, Tansella M, Petrovich L. Second-generation antipsychotics and risk of cerebrovascular accidents in the elderly. Journal of clinical psychopharmacology. 2005 Oct;25(5):468-70.
59. Barnett MJ, Wehring H, Perry PJ. Comparison of risk of cerebrovascular events in an elderly VA population with dementia between antipsychotic and nonantipsychotic users. J Clin Psychopharmacol. 2007 Dec;27(6):595-601.
60. Sacchetti E, Trifiro G, Caputi A, Turrina C, Spina E, Cricelli C, et al. Risk of stroke with typical and atypical anti-psychotics: a retrospective cohort study including unexposed subjects. J Psychopharmacol. 2008 Jan;22(1):39-46.
61. Layton D, Harris S, Wilton LV, Shakir SA. Comparison of incidence rates of cerebrovascular accidents and transient ischaemic attacks in observational cohort studies of patients prescribed risperidone, quetiapine or olanzapine in general practice in England including patients with dementia. J Psychopharmacol. 2005 Sep;19(5):473-82.
